(1990-07-11) Deming talk Does anybody give a hoot about profit?

Deming talk: Does anybody give a hoot about profit? On the afternoon of 11 July 1990, Dr W Edwards Deming gave a short presentation to some 25 executives from major European companies. This article provides an edited transcript of that presentation and the subsequent Question and Response Session. The presentation was structured around Dr Deming’s paper “A System of Profound Knowledge” (May 1990).

Think what you could do for the world -- provided what? Provided you knew what to do. How would you know? How would you know?

We’re talking about quality. Everybody has the answer: everybody knows just what to do.

I decided some two years ago on Deming’s Second Theorem: “We are being ruined by best efforts.” Ruined by best efforts, doing the wrong thing, with hard work,

...someone enquired of me: “You’ve told us Deming’s Second Theorem. But what is Deming’s First Theorem?” My reply was that I hadn’t thought what it should be. But I began to worry about it. And during the night, a few weeks later, I decided. Deming’s First Theorem is: “Nobody gives a hoot about profit.”
...I mean long-term profit (sustainable)

(I am, incidentally, assuming that there’s no harm in profit.)

Where does the power of a leader come from? It comes from knowledge, I believe. He has power by virtue of his position, of course. But stronger power comes from knowledge.

Let’s think again of those wrong answers on how to achieve quality. Some of them are positively harmful. The worst one of all is ranking -- toughen up the annual appraisal of people! The ranking of people, teams, plants, divisions, with reward and punishment, is wrong. What’s wrong? Failure to appreciate variation: fundamentally, that’s what’s wrong. The source of the problem is failure to understand a little bit about variation; just a little bit would do.

What’s missing here, from all this? There is a missing ingredient. And I have no name for it except Profound Knowledge.

The System of Profound Knowledge appears here in four parts:

• Appreciation for a system: (see below)
• Some knowledge of variation: In an hour, I can only use words without any explanation and hardly any examples: there’s not time -- it’s an impossible task.
• Theory of knowledge: Theory of knowledge: by which we know that experience, in the absence of theory, teaches nothing... An example teaches nothing. Suppose you study a company. You’ll learn nothing without theory; for all people can do then is copy. If we have some theory (hypothesis), then that theory leads to questions, by which we learn. Either the theory seems to hold, or we may have to abandon it, or modify it.
• Psychology: Psychology is an important ingredient, I think, of this System of Profound Knowledge. For example, if psychologists understood variation (only just a little bit, what we learn from the Experiment with the Red-Bead Experiment [3, page 3]), they would never again participate in improvement of instruments, questionnaires, or other ways to carry out annual appraisal of people. (Unless there's money in it)

If economists understood a system, the theory of a system, and the role of cooperation in optimization (which I’ll come to), they would no longer teach and preach salvation through adversarial competition. They would instead lead us into optimization, by which everybody would come out ahead. (Central Planning?)

I’d like to talk about a system. (I’ll call it a system: you may have another word for it.) What do we mean by a system? I’ll define a system as a network of components, parts of the Organization, functions, activities, that work together for the aim of the Organization. There must be an aim. (hmm maybe Purposeful System? see (1995-12-31) Purser On Defining Systems)

Remember Deming’s First Theorem: does anybody give a hoot about profit? I think that, within 30 minutes, you’ll agree with me that nobody really cares, not really, not enough to do anything about it. Worry, yes; best efforts and hard work, yes. But best efforts and hard work do not produce desired results without knowledge; there is no substitute for knowledge. A flow diagram will help; and it doesn’t need to be too complicated.

Figure 1 is a flow-diagram of a system. I will take manufacturing as the example,

we do consumer research to try to discover what improvement or innovation in product or service might help the customer in the future, and will entice him to buy. That may call for different inputs, and design or redesign of product or service.

I believe that this diagram (Figure 1) made the difference in Japan. I’d been to Japan twice, in 1947 and 1948, at the invitation of one of General Douglas MacArthur’s people

I was able to go in June 1950, and I was there through July and August.

It is important to see that top management invited me: Mr Ishikawa was also President of the great Kei-dan-ren, the association of top management. He was highly respected, unselfish, brilliant. He invited me, and I took that to mean he was interested. He was. I knew that I would accomplish nothing unless he brought top management together. He did so, at a dinner meeting in the Industry Club in Tokyo.

If you read the report written by the Japanese, I think you will find that the greatest way I accomplished anything there was through that diagram, Figure 1 -- and the fact that I had faith in them. I knew that they could do it. I told them that I thought I knew what they should do, and that I could help. I was back in six months, back in another six months, back the year after that. Every one of the 21 people in top management was there every time, at every conference.

I told them that, within five years, manufacturers the world over would be trembling and would begin to scream for protection. Several people in top management told me later that I was the only man in Japan who believed it. They beat it by a year.

Figure 1 was the most important force, because through it they saw manufacturing as a system, the whole operation as a system. They already had knowledge, great knowledge, but it was in bits and pieces. Some of the bits and pieces were in conflict with each other. They had knowledge; but they did not see it as a system.

The aim of a system is not a theorem: you don’t derive it from axioms and corollaries. The aim that I’d propose for a system is gain for everybody --stockholders, employees, suppliers, customers, the community, the environment. Everybody should fare better, everybody in the system should be ahead, his quality of life should improve: that’s what I mean by the aim of a system. (cf World Game)

And the aim of the system must be clear to anybody in it. If we understand a system, and work on it, then everybody will gain. (Positive-sum game)

Optimization of a system should be the basis for negotiation between any two people. They should form a system. Any two people should regard themselves as components of a system; any two divisions, or three, union, labor, and management, should regard themselves as parts of the same system. And the aim is optimization, in which everybody gains. This should be the rule for negotiation between competitors, between countries. Everybody would gain.

Let’s look at the diagram in Figure 3. It has a scale showing degrees of interdependence: low at the left, high at the right

A bowling-team has low interdependence (while they are playing): the team is rated just by adding up the scores

We have on the right of the diagram examples of high interdependence: a business and an orchestra

what is the difference between the Royal Philharmonic Orchestra of London and some other orchestra that is not as good? Same music, same specifications, not a mistake from either orchestra, everything just right. What’s the difference? Just listen to the difference. One is managed better than the other. The players of the Royal Philharmonic Orchestra of London are there to support each other: every one of the 140 supports the other 139. They’re not there as prima donnas.

You could have the best engineer, the best man in finance, the best man in manufacturing, the greatest man in distribution, in all this world: but you would not have a company -- unless they work together. Having the best is not the answer. Sure, they must have some competence. But the answer lies in management, and whether people understand their jobs as part of a system.

The conductor of the orchestra is a manager

The performance of any component is to be judged in terms of its contribution to the aim of the system, not for its individual production or profit, nor for any other competitive measure. Think of the companies that are dashing themselves on the rocks by introducing competition between their components.

Just an example: It would be poor management to save money on traveling expenses without regard to the physical welfare of the travelers... Can you blame the Travel Department for saving money when that’s their job? That’s their job: to save money -- never mind the traveller. The Travel Department will get a plus, but the traveller six minuses, for she would be totally unfit for her job.

Delegate: Dr Deming, could I ask you a question? Taking your system model of a manufacturing business, could I ask you your views on introducing the owners’ of the business into the system? What lies behind my question is that, many of us believe, we in Western manufacturing are at a disadvantage because the job of the analysts in the City of London, and the other stock exchanges of the West, is to maximize short-term (speculative) profits.

Deming: Short-term: that’s the problem. May I put it along with some other problems?

let me try to put this together. Consider somebody (or a company) who decides to become selfish, only concerned about himself -- see Figure 4. It’s impossible. He cannot do this, he cannot draw a ring around himself.

Whether you like it or not, they are all part of the system, and we need to optimize the whole system. Now, the bigger we make the system (not only this company: let’s include suppliers and customers), the better will be the chance for greater gain to everybody; but also, the more difficult is the problem of management. Does this command and control really scale?

As we expand, as we include more activities and more companies, the more difficult it is to manage, and the more knowledge is required for management of the system. How far can we expand? Let’s take short-term thinking, and some other evils in the system. Let’s talk about them separately. In the first place: it is wrong, a wrong theorem, that “if you cannot measure it, you cannot manage it”. The most important losses and gains cannot be measured, yet they must be managed.

Let’s write down here some of the gains and some of the losses. Gains from: · long-range planning · education · training · morale · loyalty

Losses from: · short-term thinking · emphasis on short term gains · ranking (people, teams, plants, divisions) · annual appraisal of people · bonuses · incentive pay · MBO · Quotas for production, for sales, · uncoordinated business plans

These are heavy gains and losses: their magnitude unknowable, most of them not even recognized, not even under suspicion; these are the ones we must manage.

Think of the losses. Think of the mismanagement that I’ve heard about, some that has taken place just in the last six months

they become competitive: anything to get more money

Why introduce conflict? Conflict means we break up the system into components, and the components then become competitive, whereas they should be managed for optimization.

Some of the corollaries are frightening. Is a popular vote going to make things better? How could it, without knowledge?

Question and Response Session

Delegate: What are the reasons for the Japanese readily accepting your approach of optimization of the system, whereas the Europeans and Americans are struggling and not succeeding with it? Dr Deming: A Japanese is never too old or too successful to learn. They’re in a learning mode all the time. That’s the only explanation I can offer.

Some people say that they listened and learned because they were in a crisis. We’re in a worse crisis, for it’s not visible. A fish does not know that it’s in water: it doesn’t give a hoot about water

Delegate: How does Genichi Taguchi’s philosophy fit into your own theory? Dr Deming: It’s all I’ve been talking about

The Taguchi loss function will be a parabola at the bottom. It can be steeper on one side and not so steep on the other side. But the two halves will be parabolic at the bottom. So we don’t need to have precise optimization.

But when we fail to do that, we lose -everybody loses. That was Taguchi’s theorem. I was there, in September 1960, when he read his paper in Tokyo. This is exactly what I’ve been talking about

Delegate: The Japanese system is successful, and it is based on competition and ranking ... Dr Deming: It is not! No, this is misunderstanding. The Japanese system is based on cooperation, not on ranking. Read the paper by Dr Yoshida [6].

Do you know the difference in pay between somebody that went through university in Japan and somebody that didn’t get into university? It’s something like three dollars a month. It’s trivial. (income inequality)

Sure, the Japanese have examinations in education. But a Japanese child is never humiliated -- never, either at home or in school. His teachers are supportive; the Japanese support people. That is why, when a Japanese child fails, he feels he’s let down his family, he’s let down all his friends, he’s let down his teachers. That’s the reason for their high suicide rate. That’s not good; no, no. But it’s because people supported him: he feels he’s let them down. This feels like some asia-phile rose-colored glasses.

Delegate: But, once you benchmark, you automatically make a ranking. For instance, we have several factories and, in order to give a positive aim, a target, we issue a kind of ranking which is a benchmarking. I can’t see any negative approach in that. It is being positive. It is a kind of positive race toward a target.

Dr Deming: No, it is not. Of course it’s not. It’s destructive -- that’s my message -- destructive. Read the paper by Nolan and Provost [7]. It’s a matter of understanding variation

Delegate: I understand that benchmarking is one of the tools developed in Japan by Tsuda. (I ask; I am not sure.)

Dr Deming: I don’t believe it. I think this is a misunderstanding. I know him well: he’s a brilliant man. I don’t think the Japanese do that. How could they? How could they? They’re supportive

Chairman: I think, Dr Deming, it may be that benchmarking is used not to rank but to see where you might improve. Certainly some people, as Peter Bonfield explained, use benchmarking to find out what is the best their competitors have achieved in various areas, so they learn what is the gap between what they’re doing and what they might do.

**Dr Deming: Well, I think it’s a waste of time. In any case, there are some things you can’t benchmark. But that’s not what I’m talking about here. Think of the man-years that are wasted in worrying about competitors, getting figures on the competition, showing trends -- the competition going this way, we going that way. Think of all the time lost. All that effort could be put to better use. People worry about the competition. Better to forget them; they have their own problems. And the competition is part of the system: what is best for you is best for them. (grow the pie)

Delegate: If MBO is bad, should a company set a budget?

Dr Deming: They’d damned well better set a budget! But MBO asks somebody to do what is outside the system. It comes from failure to understand a stable system... MBO consists of asking for the point, the target, which may well be outside the system. Anybody can make it happen. He’d better make it happen, or he won’t have a job. He makes it happen -- at impairment or destruction of the company in some other part.

**If you want this result, the only safe way to get it -- the only way you can afford to do it -- is to improve the system so that you include this result. This is management’s job. It’s all right to have a numerical goal, but only the method will get you there: it’s only the method that counts. Anything else will mean loss... His job is to try to discover who is outside the system (outside the normal variation), in need of special help -- and he has the figures to do it with.